跳到主要內容

臺灣博碩士論文加值系統

(44.192.22.242) 您好!臺灣時間:2021/07/28 06:40
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

我願授權國圖
: 
twitterline
研究生:葉晉逢
研究生(外文):CHIN-FENG YEH
論文名稱:應用相互訊息模型於文章推薦之研究
論文名稱(外文):The study of applying mutual information model to document recommendation
指導教授:翁頌舜翁頌舜引用關係
指導教授(外文):SUNG-SHUN WENG
學位類別:碩士
校院名稱:輔仁大學
系所名稱:資訊管理學系
學門:電算機學門
學類:電算機一般學類
論文種類:學術論文
論文出版年:2009
畢業學年度:97
語文別:中文
論文頁數:76
中文關鍵詞:文字探勘相互訊息向量空間推薦機制
外文關鍵詞:Text miningMutual InformationVector Space ModelDocument Recommendation
相關次數:
  • 被引用被引用:0
  • 點閱點閱:204
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
網路資訊型態、內容不斷演進,網路豐富的資訊已成為大眾在資料蒐集上的使用基礎,在豐富資訊可供使用的背後,同時也產生了資料搜尋的問題。早期的資料搜尋提供的是「量」的層面,使用者考量的層面是資料多不多、廣不廣,而今過多的資料卻反而造成了搜尋上的問題,使用者搜尋行為也從「量」的搜尋轉變成「質」的搜尋,所謂「質」的搜尋,也就是尋找的資料是不是符合使用者預期的有效資訊,而現今的推薦系統,也就是在提供解決搜尋時「質」的問題,透過推薦系統,輔助使用者進行搜尋資料篩選,找到可用的合適資料,傳統利用空間向量模型的推薦系統,在某些情況下會造成資訊上的盲點,同時在大量資料下會產生系統效能的問題,本研究希望能找出其他的推薦機制,用以解決目前的不足及改善之處。
本研究中提出以相互訊息模式應用於推薦系統的概念,藉由維基百科中的文章間的連結(Link)共同出現的頻率作為推薦系統的分析基礎,利用相互訊息模型提高推薦系統的表現。目前相互資訊模型大多應用於其他面向的研究,故本研究也希望評估該模式是否適用於資訊的推薦。
經過實驗結果顯示,相互訊息應用於維基百科資料庫中,有不錯的推薦表現,平均的推薦精確率可達70%,與傳統的空間向量模型比較之下,在各項參數值的交互測試中表現更為穩定,故本研究所提出之相互訊息模型確實可應用於推薦機制上,同時亦能改善傳統推薦機制的不足之處。
There has been a revolutionary change in the way people collect information. Information comes mostly from searching on the Internet and there is a variety of information available on the Internet. The Internet has been one of the fundamental sources of data collection. The revolution has brought up several problems when collecting information on the Internet. In earlier days, people are looking for the “quantity” of information collected from the Internet, caring about whether data sources are as many as possible. Nowadays, the trend is changing to search the “right” content as “quality-oriented” search. The quality-oriented search focuses on the “effectiveness” of the content and cares about whether the information is matching the expectation of the data collectors’ needs in context. The “recommendation agent” is targeting to assist people to filter the information for those better fulfills the queries. The traditional mechanism for “recommendation agent” is to leverage “Vector Space Model” algorithms to obtain the co-relationship between content sources. However, there are certain limitations behind this kind of algorithm. At the same time, it is system resource consuming for running vector space model programs and this further limits the applications of such algorithm. In this thesis, it is aimed to find other recommendation mechanism to complement the searching tools on the Internet.
In this study, “Mutual Information Model” is investigated to recommend relative documents by using frequencies of links between documents in WiKipedia as the basis for co-relationship analysis. Mutual information model is used to strengthen the performance of documentation mechanism. At the moment, mutual information model is mostly applied in other perspectives. Therefore, it is aimed to investigate the appropriateness of applications of this model on document recommendation.
According to the experiments for this study, applying mutual information model on Wikipedia database has generated satisfying results with precision rate of more than 70%. Compared with traditional vector space model, the performance of mutual information model is rather stable in the mutual testing with different parameter settings. Therefore, it is concluded in this study that it is applicable to apply mutual information model on document recommendation with certain improvements from the traditional document recommendation mechanism.
目 錄
表 次 vii
圖 次 viii
第壹章 緒 論 1
第一節 研究背景與動機 1
第二節 研究目的 2
第三節 研究架構 3
第貳章 文獻探討 5
第一節 推薦系統 5
第二節 特徵選取 7
第三節 關聯性計算 12
第四節 連結 16
第三章 研究方法 17
第一節 問題描述 17
第二節 系統架構 18
第四章 實驗結果與分析 29
第一節 實驗的資料來源 29
第二節 實驗的流程 32
第三節 實驗的評估方式 46
第四節 實驗的結果及分析 48
第五章 結論與建議 68
第一節 研究結果 68
第二節 研究貢獻 70
第三節 後續研究建議 71
參考文獻 72
參考文獻
一、中文部份
1.吳紫葦,Extraction of Multiword Expressions related to Grammatical Collocation Based on Syntactic and Statistical Information,國立清華大學資訊系統與應用研究所碩士論文,2006。
2.林正偉,Web2.0環境下以大眾分類法為基礎之知識收藏推薦系統,國立高雄第一科技大學資訊管理研究所碩士論文,2008。
3.鍾政憲,以代理人社群為基礎的主動式知識服務推薦系統之研究,大葉大學資訊管理學系碩士班碩士論文,2004。
4.張慧羚,應用本體論網路分析於資訊推薦之研究,輔仁大學資訊管理系碩士班碩士論文,2005。
5.林于荏,以多重概念之潛在語意索引進行文件比對,輔仁大學資訊管理系碩士班碩士論文,2002。
6.陳盈如,稅務法規問答系統之研究。銘傳大學資訊管理學系碩士論文,2004。
7.曾元顯,關鍵詞自動擷取技術之探討,中國圖書館學會會訊,5卷,3期(106),1997。
8.李金男,應用知網知識庫於國小課本理解之研究,國立成功大學資訊工程學系碩士論文,2000。
9.李坤霖,網際網路FAQ檢索中意圖萃取及語意比對之研究,成功大學資訊工程學系碩士論文,2000。
10.陳如妙,應用文件探勘技術於FAQ系統之建置,銘傳大學資訊管理學系碩士論文,2002。
11.中央研究院中文詞庫小組,「CKIP中文自動斷詞系統」,中央研究院,1999。
12.陳意芬,概念式自動問答探索,國立交通大學資訊科學系碩士論文,2002。
二、英文部分
1.Choueka, Y., Klein, S.T. & Neuwitz, E., Automatic retrieval of frequent idiomatic and collocational experssions in a large corpus, Journal for Literary and Linguistic Computing, 1983, pp.34-38.
2.Church, K.W. & Hanks, P., Word association norms, mutual information and lexicography, Computational Linguistics, Volume 16, No. 1, 1990, pp.22-29.
3.Cooper, R.J. & Ruger, S.M., A Simple Question Answering System, The Ninth Text REtrieval Conference (TREC 9), 2000.
4.David, W.M., Ubiquitous Recommendation Systems, Computer Society officers, Volume 36, No. 10, 2003, pp. 111-112.
5.Dunning, T., Accurate methods for the statistics of surprise and coincidence, Computational Linguistics, Volume 19, No. 1, 1990, pp.61-75.
6.Evert, S. & Krenn, B., Methods for the qualitative evaluation of lexical association measures, Computational Linguistics, Volume 39, 2001, pp. 188-195.
7.Kilgarriff, A., An MT Lexicographers Workstation Supporting State-of-the-art Lexical Disambiguation, MT Summit VII, 2001, pp.187-190.
8.Kontos, J. & Malagardi, I., Information Extraction and Knowledge Acquisition from Texts Using Bilingual Question-Answering, Journal of Intelligent and Robotic Systems, Volume 26, No. 2, 1999, pp. 103-122.
9.Kuo, J.J., Lin, K.K., Chen, H.H., Kao, C.H. & Lin, B.S., Question Type Classification and Its Application to a Question Answering System, 2002 IEEE International Conference on Systems, Man and Cybernetics, 2002, pp. 641-646.
10.Li, S., Zhang, J., Huang, X. & Bai, S., Semantic Computation in a Chinese Question-Answering System, Journal of Computer Science and Technology, Volume 17, No. 6, 2002, pp.933-939.
11.Lin, D., Extracting collocation from Text corpora, First Workshop on Computational Terminology, 1998, pp. 57-63.
12.Lin, D. & Pante, P., Discovery of Inference Rules for 47 Question-Answering, Natural Language Engineering: Cambridge University Press, 2001, pp.343-360.
13.Mobasher, B., Dai, H., Luo, T. & Nakagawa, M., Effective Personalization Based on Association Rule Discovery from Web Usage Data, the Third International Workshop on Web Information and Data Management, 2001, pp. 9-15.
14.Paul, R & Hal, R.V., Recommendation systems, Communication of ACM, Volume 40, No. 3, 1997, pp. 56-58.
15.Pearce, D., A comparative evaluation of collocation extraction techniques, Third International Conference on Language Resources and Evaluation, 2002.
16.Ricardo, B.Y. & Berthier, R.N., Modern Information Retrieval, ACM Press, Addison-Wesley, 1999.
17.Salton, G. & McGill, M.J., Introduction to Modern Information Retrieval, McGraw-Hill, Inc., 1983.
18.Salton, G. & Buckley, C., Term Weighting Approaches in Automatic Text Retrieval, Information Processing and Management, Volume 25, No. 4, 1988, pp. 513-523.
19.Salton, A.W., & Yang, C.S., A Vector Space Model for Automatic Indexing, Communications of the ACM, Volume 18, No. 11, 1997, pp. 613–620.
20.Schafer, J.B., Recommender Systems in E-Commerce, Proceedings of the first ACM Conference on Electronic Commerce, 1999, pp.158-166.
21.Schafer, J. B. & Konstan, J. A. & Riedl, J., E-Commerce Recommendation Applications, Journal of Data Mining and Knowledge Discovery, Volume 5, No. 1, 2000, pp. 115-152.
22.Shardanand, U. & Maes, P., Social Information Filtering: Algorithms for Automating ‘Word of Mouth’, the Conference on Human Factors in Computing Systems (CHI95), 1995, pp. 210-217.
23.Tsai, C.H., A Word Identification System for Mandarin Chinese Text Based on Two Variants of the Maximum Matching Algorithm, http://technology.chtsai.org/mmseg/, 2003.
24.Yang, Y. & Pedersen, J.O., A Comparative Study on Feature Selection in Text Categorization, Fourteenth International Conference on Machine Learning (ICML’97), 1997, pp. 412-420.
25.Yajuan, L. & Ming, Z. Collocation Translation Acquisition Using Monolingual Corpora, Computational Linguistics, Volume 42, No. 22, 2004, pp. 167-174.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top